: A case of what ? Methodological lessons from a reanalysis of conflicts within Swedish Juvenile Care

“Collaboration” is generally portrayed as being beneficial to authorities, even if previous collaborative research shows that conflicts are common between authorities who are supposed to cooperate. What takes place when different actors in the collaboration meet in practice? And what is the best way to analyse this? In qualitative studies, it is often problematic to go from an exhaustive analysis of individual empirical instances to an overall picture of the context or phenomenon in which all instances taken together can be viewed as a case. Years of close engagement with the data may interfere with the analyst’s capacities and opportunities to contextualize a study more broadly and theoretically, and detailed knowledge about a range of situations in the field may make novel contextualizations difficult. This article discusses how to overcome such obstacles, using examples from a study about a “collaboration” project in Swedish youth care.


Introduction
"Collaboration" is an important aspect of social work, with previous collaborative research showing that conflicts are common between organizations and authorities that are supposed to cooperate, in addition to between collaborating actors (Danermark and Kullberg, 1999;Bergmark and Lundström, 2005;Hjortsjö, 2006; Axelsson and Bihari Axelsson, 2007), though the research does not always show precisely how this occurs. In my study, I will show how this happens by focusing on conflicts and drawing lines on social work (and organizational) practice.
When analysing qualitative data, it can be difficult to shift from an exhaustive analysis of individual scenarios to an overall picture of these scenarios together as a single case. An analyst who has spent a significant amount of time with the data might have trouble taking a step back and contextualizing the study from a broader perspective.
Moreover, detailed knowledge of the field can make deriving novel contextualization a challenge (Ragin and Becker, 1992).
Using as an example the aggregated instances from a study about a "collaboration" project in Swedish youth care, this paper examines how to overcome such obstacles.
The purpose of the project was to improve collaboration between the social services and the Swedish National Board of Institutional Care (Statens Institutionsstyrelse or SiS). The idea was that employees in a new professional category -designated as "coordinators" -would coordinate care measures concerning young people (and their families) under the care of society to ensure that officials met their commitments and also to act in the role of a state-employed parent.
This collaborative project within Swedish youth care was analysed as part of my dissertation (Basic, 2012), in which conflicts, alliances and comparisons were identified in interviews and observations. The empirical material is contained in recorded informal interviews with 147 project participants (institutionalized youngsters, their parents and numerous professional groups within the social services, the National Board of Institutional Care and the project itself), 1 as well as through observations of organized meetings, conversational get-togethers both before and after organized meetings, and visits to institutions, social service offices, the head office of the Swedish National Board of Institutional Care and MVG 2 offices (Emerson, Fretz, and Shaw, 1995;Fangen, 2005;Holstein and Gubrium, 1995;Kusenbach, 2003;Riessman, 1993;Silverman, 1993Silverman, /2006. In analysing this material, I benefited from the insights of Georg Simmel (1908/70;1950/1964, Theodore Caplow (1968) and Erving Goffman (1959Goffman ( /1990, while the empirical material was collected by three researchers (Basic, Thelander, and Åkerström, 2009).
Even though the study yielded detailed knowledge about various forms and constellations of conflicts within this collaboration project, an overall picture could be clarified only when all of these instances were reanalysed in terms of previous research on collaboration in other areas and between other organizations. Parallels and variances among the retold conflicts throughout the interviews and the interactional, in-situation-shaped conflicts showed that: (1) the "collaboration" project became a struggle in Georg Simmel's (1908/70) meaning, and that (2) the client in human service organizations now faces a notable risk of falling outside such new "collaborations". What this article seeks to specify is how this result was obtained, and what it means in practice to view a qualitative data set from a wider perspective.
My analytical findings are presented in the following themes: (1) Collaboration became a struggle, (2) Conflicts between organizations, (3) Descriptions from juveniles and parents of conflicts, (4) Meeting conflicts, and (5) The client falling Social Services staff members: 37 individuals; Institutional staff members: 26 individuals; MVG project employees (managers and coordinators): 31 individuals (six MVG coordinators were interviewed twice); HVB staff members (HVB = home for care and housing, i.e. institutions under private management not managed by SiS): two individuals. 2 MVG in Swedish is short for "Motverka Våld och Gäng" ("Counteract Violence and Gangs"), which was the formal name of the collaboration project. The number of observations and field notes, including those from the interviews, was 119. The coordinators were observed on 70 occasions, and we focused on the interaction of the coordinators (during the meeting and on a daily basis) with other actors, namely the social secretaries, treatment assistants and unit managers at the institutions, in addition to youngsters, parents, school representatives and family therapists. Official meetings in which only professionals participated were observed and recorded (we always asked the participants if they would agree to this). In such cases, we focused on the interaction between the participants of the meeting: representatives from the project management, the coordinators, unit managers from the social services and institution managers from SiS. outside collaborations. The position of my approach is well within both the relevant body of sociology (micro/small group interaction: Simmel, 1908Simmel, /70, 1950Simmel, /1964Caplow, 1968and Goffman, 1959/1990) and methodology (Ragin and Becker, 1992;Emerson, Fretz, and Shaw 1995;Fangen, 2005;Holstein and Gubrium 1995;Kusenbach, 2003;Potter, 1996Potter, /2007Riessman, 1993;Silverman, 1993Silverman, /2006Flybjerg, 2003Flybjerg, /2004. My analysis shows what takes place in the detailed interaction, which then brings a conclusion that goes counter to much organizational/political talk advocating inter-organizational collaboration. In this way, the article contributes to a sociological understanding of a contemporary widespread social phenomenon -"collaboration".

Method
The problems associated with doing and understanding case studies involve, apparently necessarily, the question of an explanation or description, which might be translated as the problem of what we can say about what we've found out in our research. Can we say that something we discovered causes or produces or influences or comes before or in some other way affects what happens to some other thing? (Becker, 1992: 205) According to the conventional positive view, a case study cannot be valuable if it is not linked to a hypothesis that tracks the well-known hypothesis of a deductive explanatory model (Abercrombie et al., 1984: 34;Campbell and Stanley, 1966: 6-7).
Instead, Ragin and Becker (1992) propose that the researcher who carefully examines individual cases should not aim to strengthen something, rather than learn and understand the phenomenon. According to Wieviorka (1992), researchers who have implemented in-depth studies of cases will usually reach the conclusion that their prejudices, presumptions, concepts and hypothesis are incorrect, and that the material from the case would have forced them to revise their hypothesis in essential aspects. Ragin (1992: 225) designates this phenomenon as a "special feature of small-N research", and explains that criticism of studies based on a single case is in many occasions misguided because the analytical findings in case studies can be linked in many different ways to an overarching echelon. Ragin (1992: 225) writes: … small-N qualitative research is most often at the forefront of theoretical development. When Ns are large, there are few opportunities for revising a casing. At the start of the analysis, cases are decomposed into variables, and almost the entire dialogue of ideas and evidence occurs through variables. One implication of this discussion is that to the extent that large-N research can be sensitized to the diversity and potential heterogeneity of the cases included in an analysis, large-N research may play a more important part in the advancement of social science theory.
As mentioned above, this difference between large selections and individual cases can be figured out in terms of the importance of human learning. If you assume that the researcher's target is to understand and learn about the studied phenomenon, then research is simply a form of education. If you assume that research, like other learning processes, can be described with the phenomenology of human learning, it will be clear that the most advanced form of understanding will be achieved when researchers place themselves within the studied context, and only in that way can researchers understand the social participants' speech, writing, actions and interaction (Flybjerg, 2003(Flybjerg, /2004. Flybjerg (2003Flybjerg ( /2004, who expands on the thoughts of Ragin and Becker (1992), argues that there are five common misunderstandings about case-study research.
The first misunderstanding is that theoretical knowledge is more valuable than practical knowledge, and Flybjerg (2003Flybjerg ( /2004 revises the misunderstanding by arguing that predictive theories cannot be found during studies on human relationships. Concrete, contextualized knowledge is therefore more valuable than a vain search for predictive theories. The second misunderstanding is that one cannot generalize from a single case; therefore, the single case study cannot contribute to scientific development. Flybjerg (2003/2004: 189-193) corrects this second misunderstanding and writes that you can often generalize on the basis of a single case, and that case studies can be central for scientific development through generalizations as supplementing-or as an alternative to other methods. Flybjerg means that formal generalizations are overvalued as a source for scientific development, while the power of an example is underrated.
Journal of Comparative Social Work 2013/2 8 The third misunderstanding is that the case study is most useful for generating hypotheses, but that other methods are more suitable for hypothesis testing and theory building. With a starting point in the second revised misunderstanding, Flybjerg (2003Flybjerg ( /2004 corrects the third misunderstanding and notes that the case study is useful both to generate and test a hypothesis, but is not limited to only these research activities.
The fourth misunderstanding is that the case study contains a bias towards verification, i.e. a tendency to confirm the predetermined ideas of the researcher. Flybjerg (2003Flybjerg ( /2004 instead argues that the case study does not have a greater bias towards the authentication of the researcher's predetermined notions than any other method. On the contrary, experience demonstrates that the case study has a bigger bias towards the falsification of predetermined ideas than to verification.
The fifth misunderstanding is that it is often difficult to summarize specific case studies. According to Flybjerg (2003Flybjerg ( /2004, this difficulty can especially arise in summarizing case studies, particularly those about case processes, but is less present for evaluating effects on a case. However, the problems with summarizing case studies can frequently depend on the studied properties of the reality, rather than on the case study as a research method. It is often not desirable to summarize and generalize case studies, and good studies should be read as whole narratives according to Flybjerg.
The qualitative study can be judged by its appropriateness for analysing phenomena and tendencies through its application to other and similar fields, including those not included in the present study. The point is that the reader should assess the study based on its transferability (rather than traditional generalization) by trying out the sustainability of the analysis through application to other potential analyses (Becker, 1970: 41-42;Fangen, 2004Fangen, /2005. During my dissertation work, I was influenced by the above-described perspective (Basic, 2012). According to these principles, my analytical discoveries were transferred from an analysed empirical example (interview quotes or field notes) to findings linked to a case (project), and then retransferred to analytic findings in a more overall collaboration context, as represented by previous research conducted on the social phenomenon of "collaboration".

Analytic starting points
This article joins the narrative traditions within sociology that consider oral descriptions as both experience-based and discursive (Potter, 1996(Potter, /2007. In addition to this general point, I found conflicts to be particularly relevant components of the specific stories I examined. According to Georg Simmel (1908/70: 10-21, 41-43, 117-118), in contrast to perfunctory perceptions that disagreement does not have to be viewed as disruptive to relationships, it can instead be viewed as unity and integration. He states that conflicts are an expression of the parties' solid involvement in a situation in which they share an integrated function among the actors (Simmel, 1908/70: 12-16;41-43).
Simmel sees the conflicts in the relationship as cohesive, inclusive and unifying rather than as disruptive, but to achieve this characterization requires a shift between conflict and consensus, as well as between belligerence and striving for calm and harmony in the relationship among actors (Simmel, 1908/70: 10-15, 117-118).
When Simmel (1908/70: 21, 41-43) writes about disagreements, he talks about conflicts among individuals, organizations and states in which elements of the disagreement unite these actors. Thus, conflicts are relationship-related, integral and intense in small groups; the more connections among people, the more intense are the conflicts.
According to Simmel (1908/70: 90, 110-118), disagreements can hold struggling parties concentrated on one point of interest. He means that a focus on these "conflict points of interest" enables contradiction in the same way that a lack of focus or removal of conflict items attenuates contradictions. Simmel (1950Simmel ( /1964) also analyses the relationships among actors in a triad insofar as the triad can appear in different constellations, with three individuals and the possible relationships among these individuals comprising different triad constellations. Relationships among the participants in the triad easily contribute to the exclusion of one of the participants -an alliance between two participants is created as an antipole towards the third one.
According to Caplow (1968), who used Simmel's way of thinking in analysing organizations, relationships in the triad are unstable and variable; Caplow notes that alliance constellations can shift during the interaction. A participant in a triad may feel left out, but when circumstances change, one or both of the alliance partners can feel like an intruder (Caplow, 1968: 19). This variability often makes two members of a triad form an alliance against the third, although the specific alliances can vary.
According to Simmel and Caplow, the usual pattern is that alliance formation occurs in struggle situations, and that in these situations the struggling actors can act strategically. Goffman's (1959Goffman's ( /1990) analysis of the team has also been important for my investigations, as team members tend to support one another's appearance and form a common front towards the audience. Much like the single individual, they attempt to convince their audience, define the given situation and convey their definition to the audience. Even so, problems may arise if partners in a team come to a disagreement; the appearance may then fail and disorder arise. Coincidences in the ongoing interaction therefore allow the participants of different social groups to unite in one of many temporary teams.
Conflicts and alliances are a comprehensive and exciting theme of my analysis. The perspectives of the aforementioned theorists are useful for my ambition to understand the empirical example (interview quotes or field notes), and through this to understand the social work practice, both as an analytical starting point and as a subject of nuances.

Collaboration became a struggle
One of the aims of the youth care project was to enhance the collaboration between the Swedish National Board of Institutional Care and the social services to help improve the support given to children in social custody and to their parents (SiS, 2006: 1-3). Paradoxically, the most salient result (Basic, 2012: 68-75) was that tensions between these two organizations and relevant actors came to the surface.

Conflicts between organizations
It clearly emerges in the empirical material that some of the employees of the social services and the National Board of Institutional Care were questioning the project, claiming their organizational mandate and obtained information, as well as applying interaction skills with the youngsters.
In these formulations, hostility appears in relations between the categories of "social services" and "government". In a conversation about why the project was started, a unit manager of social services and a social secretary relate the following: Petra (unit manager): This entire situation is about an attitude conflict between the state (government) and municipalities, and it is a shame that social workers on the lower level had to take on the burden (the conflict). Ewa (social secretary): And they (who started the project) tell us that we are doing a bad job … that it is a bad start to the project.
The government started this project through the practical actions of the National Board of Institutional Care. Petra, a unit manager of social services speaks in a way that the category of "state" gets attention, even though government representatives are not present in the project. Representatives of the government are not active, and when Petra says "state," she seems to mean the National Board of Institutional Care (or the project leaders). The state in the sequence above does not occur as an actor separated from the National Board of Institutional Care, but as a contrast to municipalities.
Petra and Ewa construct the National Board of Institutional Care (controlled by the state) as the state's alliance partner. The municipality is described as the third part in the relationship, and the category of "social workers on the lower level" is portrayed as part of a victim category in this fight; they are the ones who will push the project forward ("had to take on the burden"). Conflicts between the National Board of Institutional Care and social services were also observed in the interaction, and not only in the conversations. At various gatherings, government representatives were sometimes transformed visibly and in solid terms. This tendency can be illustrated by events at a lunch for a meeting: … at the premises, there are about seven unit heads from the various municipalities. /.../ The meeting host will arrive and show us the restaurant where lunch will be eaten. Two tables are reserved by SiS. We who have come on time (12:00)  These field notes from the restaurant give us a picture of the conflicts between the National Board of Institutional Care and social services, as the unit manager of social services sat at the table during lunch talking about the others before and after they came to the restaurant. The distance between the groups was noted when representatives from the National Board of Institutional Care purported to be slightly uninterested by lunch. Vanja comments, "If they wanted to, we could sit together, as with those on the other side," and she seems to show an interactive desire to socialize with members of the other organization. This manifestation can be interpreted as a possible invitation to reconciliation after the ease of conflicts that arose at this table (Simmel, 1908(Simmel, /1970.
The aforementioned presentations and circumstances portray a disagreement concerning the relations between the dyad of "social services" and the "state". The alliance between the "state" and the Swedish National Board of Institutional Care is often presented as the larger party in the relationship with the social services, whereas the latter is a municipality organization, and considered a lesser party.
However, the actors are interactively described as cooperating parties applying a joint effort -the project (Basic, 2012: 67-108;Caplow, 1968;Goffman, 1959Goffman, /1990Simmel, 1908Simmel, /1970. Conflicts are reflected not only in the inter-organizational relationship among the National Board of Institutional Care, social services and projects, but also in interpersonal interactions between individuals working around young people and their parents. When a treatment assistant told me about how the coordinator "took over his meeting", he was clearly upset. At this meeting, the treatment assistant wanted to discuss with social services whether or not they had arranged school and recreational activities for a boy, Fadil, who would soon move to a foster home. The coordinator preferred to discuss the points on the "agreement", which is the coordinator's written document, and already being used in other models of care such as the BBIC 3 and the ADAD interviews. 4 During the interview, the treatment assistant became upset several times, turning red in the face and raising his voice when talking about meeting events. According to him, all attendees except for the coordinator Tobias -i.e. the treatment assistant himself, the social worker, the boy Fadil and his mother -were disappointed about the coordinator's actions during the meeting (field notes). The interview shows that the treatment assistant, Allan, often hesitates in his speech, pausing frequently, and that these breaks are reported in seconds in parentheses. Additionally, he sometimes speaks with a lower or higher voice, which is marked in the text by ° for a lower-and uppercase letters (e.g. NO) for a higher voice.
Allan: Actually, we shall discuss the formalities surrounding his (Fadil's) moving out, when he is moving, eee (.), how it will be done (.) eee, how the mother is feeling and so on, and so on. Goran (interviewer): Mmm. Allan: But then, Tobias approaches with his paper (agreement) and says "NO". Then he takes over the entire meeting, now this is what happens today (hitting the table three times with his hand), and (3) I am damn frustrated with it. I thought, "HELL, it has nothing do with it." / ... / He took over the meeting, everything revolved around this as well. / ... / and I, I was so damn pissed because of it / ... / I was totally unprepared, I sat there like a question mark. °What the hell?° During the narrative, the treatment assistant Allan is engaged and upset, quoting coordinators who were involved in Fadil's case. Allan's performance strengthens the image of a dedicated narrative-telling, as he gives life to the story through both letting the character "coordinator" speak loudly for himself ("NO") and citing his own thoughts ("HELL, it has nothing do with it"). By citing the involved coordinator, Allan is distancing himself from his actions, thereby inserting the picture of a special coordinator who emerges as a bureaucrat who took over the meeting, and who fights for an administrative tool and not for the youngster.
In my previous analyses (Basic 2012: 67-108), the categories of "coordinator", "project" and "agreement" emerge in described and observed conflicts. These categories are used as struggle requisites when actors emphasize the competence of their own organization and when they defend their organizational turf.
Struggles between different officials are occasionally described with intensity, and the motive for the involved critiques may be the size of the national youth care; this world is rather small and limited. The officials in this study were often familiar with each other or at least had heard about each other, which may have caused more powerful antagonism than if they had been strangers. In the material, there are few commendatory stories except when approval is directed towards the narrator himself (Basic, 2012: 67-108;Simmel, 1908Simmel, /1970.
Showing that a conflict is a procedure of social interaction aimed at persuading those present about the veracity of the formed reality, the tools used for this purpose are adaptable and rhetorically powerful. For example, when a treatment assistant narrates how a coordinator took over a meeting using an on-paper agreement, he does so with great commitment. He passionately shares his feelings and animates the role of "coordinator" by citing him and banging his fist on the table. With this display, he is presenting his identity as a care-giving practitioner (Basic, 2012: 67-108;Potter, 1996Potter, /2007.

Descriptions from juveniles and parents of conflicts
Youngsters and parents often fight for fairness after being abused. On the one hand, they denounce the actions of the officials as being morally incorrect, while on the other, they attempt to spin the situation into something more positive for themselves, e.g. by attempting to move into alliances with officials (Basic, 2012: 147-186;Caplow, 1968;Goffman, 1959Goffman, /1990Simmel, 1908Simmel, /1970. According to the project description, the coordinators expected that they sometimes would act as meeting chairman (National Board of Institutional Care, 2006). Nonetheless, the title of "meeting chairman" was sometimes difficult to achieve by single coordinators, and their attempts to acquire this title created conflicts. The coordinator's claim to the title of chairman of the meeting could disturb the status quo and affect the actors' professional identities, hence leading to the situation arising later and being reproduced as a conflict.
Safet, an institutionally placed boy, describes a conflict that arose during a meeting.
On one side were the department director Freja and the social secretary Conny; on the other side was the coordinator Carita. A discussion among Carita, Conny and Freja is retold through quotes, and in this way, Safet demonstrates a conflict among these three: Safet: They (department director, Freja, and social secretary Conny) have said to Carita (coordinator), "so why ARE you the host of this meeting?", you know MVG is supposed to host it because they send out forms and stuff, °then°, my soc (Safet means social secretary), she told Carita, "ARE you the host of this meeting" and "who gave you the rights to do so" and stuff. So you know, it became a little (.) commotion there /.../ My soc told Carita, "you're not in charge or anything, you're just here or what?" Carita said, "Yes, I'm just here for the most, I write and stuff," then she said, "You're not making any decisions," so THAT became a little bit like that, you know. They-they-THEY wanted them to determine.
We can read that Safet is critical of Freja's and Conny's treatment of Carita, and that he is portraying a conflict by narrating the occasion. A coordinator-excluding alliance between a social worker and a department director becomes evident when a youngster tells of a conflict. Coordinator-excluding alliances propose that the officials have concealed interests, such as striving to work in an arranged fashion, appreciating peace and quiet at work and working without having to alter any habits.
Carita is left outside because she shook up the situation when she claimed the title of "meeting chairman".
In my previous analyses (Basic, 2012: 109-145), the narrated battles in youth care were depicted by youngsters and parents. For example, the battles may be about being approved for a continued stay at the institution, being able to be present at a meeting or being able to obtain compensation after being physically and mentally abused at the institution. In the recounted conflicts, there are portrayals of: (1) different alliance constellations, and (2) different roles in changed constellations, with 13 portrayed alliance constellations analysed. The classification above is mostly instituted based on four premises, specifically the temporal development of the alliances in the triad, the including and excluding function of the alliance in the triad, the influence of the alliance on the hierarchy in the triad and the effect of the alliances in the triad. In the analysis, parts and alliances are brought together, and it is revealed here that a modification of roles may bring about a variation in alliances (Caplow, 1968;Goffman, 1959Goffman, /1990Simmel, 1908Simmel, /1970.

Meeting conflicts
Loyalty alliances affect the alliance member's solidarity in relation to other members of the triad, and these groupings may either be expected or unexpected by the contributing actors. These alliances indicate the many dissimilar beliefs that the partakers have about each other, and when these expectations become fulfilled or unfulfilled, they can offer fuel for a hostile situation. An understandable example is found in the unexpected alliance between someone from the project management and a unit manager of social services, which excludes a coordinator. The coordinator had expected an alliance with the project management; when this alliance was unfulfilled, the result was mostly strong disagreements.
Alliances with an articulated consensus among the members are designated as coherent alliances. Examples of these alliances are found when actors mutually lay guilt on a third party, with the blame often involving drawing attention to the wrongful working practices of the coordinators, social secretaries, or institution staff members (Basic, 2012: 147-186;Caplow, 1968;Goffman, 1959Goffman, /1990Simmel, 1908Simmel, /1970. During a meeting with the coordinators and project management coordinators attending from Cimtown, they talked about their way of working. Based on the Cimtown coordinator's experiences, some principles of coordinator work could be summarized (field notes, recorded meeting). The meeting reproduced below will demonstrate that the first principle presented contains a point of interest that the social services unit managers react to. The coordinator, Thea, begins the presentation by reading the first MVG principle (youth participation), and the unit manager of social services, Ida, objects directly, turning on the word "shall". The interest point of the conflict during this meeting is the use of the word "shall", which seems to signal that the other is in control. Thea begins her presentation of MVG principles: Thea (coordinator): Youth and parents SHALL (letter A and L said extended) be there when the planning is done for the youngster (.) eee it appears actually quite often today (.), we should say, too often (.) (deep breath) that the social services hands over eee the care plan, to the institution and where they are going to do the treatment plan(.) and then it is really (laughs) difficult (laughs), so the result is wrong (said short and with strength in her voice) if I may say so. Unknown person: Mmm. Ida (unit manager): Now I am going say one thing, you can't go in here and make decisions about how others should work. It can be considerations, and you can make recommendations / ... / but you cannot go in with a "shall". (.) Just like that. Lens (project management): THAT IS CORRECT. Absolutely right. Thea (coordinator): BUT, "PARENTS and youngsters shall be involved." That's what you say to us all the time (Thea directs herself to Lens). (field notes, recorded meeting) During the continued meeting, Dessi, another unit manager of social services present, agrees with what Ida has said. The alliances that can be discerned in the conflict appear to be the coherent alliance, expected alliance and unexpected alliance of the actors.
Above, we can see that the coordinator, Thea, gets support from someone who, for me, was an unknown member of the staff. This person is not head of the unit, Ida or Dessi, or the coordinator, Thea. With a "Mmm", the unknown member seems to signal a coherence with the principles that Thea presents.
The unit manager of social services, Ida attacked Thea's presentation and the coherent alliance between Thea and the unknown member. She seems to have a will to prevent an unwished occasion through criticizing it ("you can't go in here and make decisions about how others should work"). It seems that Ida is attempting to accomplish a situation in which her own organizational decision mandate is defended.
The insertion from Lens (a person from the project management) after Ida's (head of unit) attack against Thea (the coordinator) creates a coherent alliance between Lens and Ida ("THAT IS CORRECT. Absolutely right"). Alliances that reach outside their organizational limits are identifiable in the empirical material for this study, but they are not as common as alliances consisting of parts from the same organizations.
Lens's style seems to be soft; he seems to focus on neutralizing the point of interest of the conflict -what Ida reacted to -the word "shall." He seems to try, with Simmel's (1950Simmel's ( /1964 terms, to act as a non-partisan mediator. Through his verbal insertions, Lens succeeds in making the atmosphere calm, but was attacked by a loud Thea ("But, "PARENTS … ").
The alliance between the project management representative, Lens, and the unit manager of social services, Ida, can be assumed to have been unexpected to the coordinator, Thea. What Thea expects is her own alliance with Lens. He is a part of the project management for the project where Thea is employed, and with her loudness Thea seems to signal that the team appearance she expected failed (Goffman, 1959(Goffman, /1990. In this situation, the expected alliance goes unfulfilled. The unexpected alliance between Lens and Ida seems to provide fuel for a continued battle, and Thea seems to attack the alliance between Lens and Ida with the aim of dividing it (see Divide et Impera, Simmel, 1950Simmel, /1964. In this form of interaction, the third member has an active role in trying to create a situation beneficial to himself by attempting to throw up obstacles to an alliance or among people in a category that could form a team or have the same interests. Simmel (1950/1964: 162) writes: Its outline is that two elements are initially united or mutually dependent in regard to a third, and that this third element knows how to put the forces combined against him into action against one another. The outcome is that the two either keep each other balanced so that he, who is not interfered with by either, can pursue his advantages, or that they so weaken one another that neither of them can stand up against his superiority.
With Simmel's perspective, the coordinator Thea's attack can be seen as an attempt to stave off the threat from the agreeing Lens and Ida. Thea also seems to play on the conflicts that exist between social services and the project (Simmel, 1950(Simmel, /1964. She raises a suggestion about an earlier agreement between her perspective and that of Lens ("That's what you say to us all the time"). The consensus seems to drive a block between the Lens and Ida partnership, which in turn opens up an opportunity for the unit manager of social services, Dessi, to be in solidarity with unit boss, Ida. This alliance constellation is expected (by the observer) and seems to be fulfilled.
To summarize, we can state that the alliance constellations change. The contributors enter and exit the interactive forms of collaboration to mark their places, as well as their own' and others' criticism.

The client falling outside collaborations
In the beginning of the project, numerous social processes were introduced, processes that complicate the image of simple solutions. As an alternative to enhancing the collaboration, my analysis shows that the project manifested itself and led to fights among organizations involving representatives from the social services, the National Board of Institutional Care and the project, in addition to more than a few interpersonal conflicts among representatives from the several professional categories involved, the youngsters and their parents (Basic, 2012).
Inter-organizational collaboration is frequently presented as being advantageous to organizations that are expected to collaborate, and the collaboration is completed despite uncertainty about the consequences (Lindberg, 2009: 45-51, 99).
Researchers have shown in diverse ways that collaboration points toward problems and disagreement, with Danermark and Kullber (1999: 160), Bergmark andLundström (2005), Hjortsjö (2006: 194) and Axelsson and Bihari Axelsson (2007: 23) all mentioning that collaboration between organizations can sometimes be very timeconsuming, which can also be at the cost of working with the clients. Previous research labels cooperation as something that takes place among organizations; it is not understood as something clients are directly involved in, in spite of the fact that the client's best interest is placed, rhetorically, in the centre of the activity when it is initiated. A similar common insecurity concerning the effects of collaboration are featured by Huxham and Vangen (2005: 13), who write: The overwhelming conclusion from our research is that seeking a collaborative advantage is a seriously resource-consuming activity, so it is only to be considered when the stakes are really worth pursuing. Our message to practitioners and policy makers alike is: Don't do it unless you have to. Anell and Mattisson (2009: 78-81, 85-87) claim that research on collaboration illustrates that partnership is problematic to achieve and that the research results are branded by doubt regarding the consequences for clients in care-giving organizations. They also state that actors who are expected to collaborate essentially "don't participate". Reitan (1997Reitan ( , 1998 writes that relationship establishment and sustaining relationships with other organizations always involves costs for a single organization. According to her, there is a structural unwillingness to interact with other organizations, and this constant interaction barrier is particularly visible in the human services organizations where such a client is outside its control. According to Reitan, the client is viewed in human services organization operations as a business commodity when treatment is practiced on him or her, as a production tool when he or she is expected to participate in the treatment and ultimately as a consumer of the service. Human services organizations use an unclear technology and knowledge base and diffuse objects, which leaves a big space for ideology and professional conflicts to occupy and defend the collaboration between organizations (Reitan, 1997;1998). Huxham and Beech (2008: 555-579) note that it is not astonishing that struggles constantly unfold in inter-organizational relationships in which members want to influence, control or stand against the activities of others. Collaboration among organizations goes hand in hand with conflict, says Schruijer (2008: 432), and the origins of the conflicts are usually contradictions among organizational goals, interests and identities.
On the basis of Hardy, Lawrence and Grant (2005), Lotia and Hardy (2008: 379) suggest that the interpretation of the interaction in the inter-organizational identities is the basis for a successful collaboration. They argue that the collaboration's identity is produced and reproduced in discursive practices through everyday routines, conflicts, alliance formation and joint operations. Additionally, these activities can provide a sense of belonging that is occasionally portrayed as being opposite to cooperating partners (Lotia and Hardy, 2008: 366-389). At the same time, Schruijer (2008: 433) points out that attempts to reduce conflicts between organizations do not in themselves improve or encourage cooperation among these organizations: "Conflict-reducing interventions can never in themselves stimulate collaboration." Furthermore, in my study's empirical material, there are officials who seem to fight for their organizational benefits and their own individual professional interests. In the discussions on collaboration between the social services and the Swedish National Board of Institutional Care, common inter-organizational collaboration identities seldom appear. Members of one organization tend to present themselves as being positive about collaboration in contrast to members of other organizations, and it is common that members from other organizations are said to obstruct cooperation. Simmel (1908/70: 10-15, 117-118) sees relations among actors as an interchange between conflict and consensus, and between belligerence and a striving for calm and harmony. With Simmel's intellectual proposition as a first point, we want the answer to the question: When and how is approval spoken to members of other organizations in my study?
Stories of the approval of representatives of other organizations are uncommon in my material, as in most cases praise is directed towards the one speaking the words of praise -critical voices about the others therefore take control. The stated criticism or defence against criticism is normally somewhat intense, while the narratives are enthusiastic and made trustworthy by the narrators, as if they occurred a day ago.
The storytellers define details, highlight some concepts, increase or decrease the strength of their voices and cite themselves and others (Potter, 1996(Potter, /2007. By proclaiming insight and information about detail, as well as showing emotional charge, the narrators demonstrate a partial effectiveness in using rhetorical tools in the verbal brawl and in part a form of a plan in their appearances. The point here is that appearances are used advantageously to persuade the audience. My research agrees with conflict findings in diverse contemporary Swedish collaboration projects by Hjortsjö (2006: 176-177, 190-194), Johansson (2011: 209, 217-218) and Bolin (2011: 110, 130, 156, 172-177), and then expands on the reasoning overlying these conflicts based on analyses of my own empirical examples. Linked with the mentioned researchers, other analytical lenses are used, and my empirical material is analysed in greater detail. In this way, my analysis demonstrates, among other things, how conflicts take shape and how they are characterized in the communication.
As several researchers have noted (Axelsson and Bihari Axelsson, 2007: 23;Bergmark and Lundström, 2005;Danermark and Kullberg, 1999: 160;Hjortsjö, 2006: 194), collaboration between organizations and authorities can take a considerable amount of time at the expense of working with clients. Collaboration is described in previous research as taking place between organizations, though not strictly including clients, regardless of rhetorically placing and focusing on the client's best interest. Comparable tendencies can be found in the empirical material of Basic (2012), as juveniles who were subject to the youth project took part in a forced relationship with officials who were expected to collaborate with the best interest of the youngsters and their parents in mind. These officials presented a negative image of their collaboration partners, and what impact this had on the treatment of the young people is somewhat clarified by the result of the effect evaluation of the youth project. Lundström, Sallnäs and Andersson Vogel (2012: 64) draw a general conclusion: There was no effect on youngsters participating in the project through the project's and coordinator's efforts. My analysis shows that when a new partnership project is launched and organizations are brought together, cooperation has a habit of becoming a struggle, and neither inter-organizational nor interpersonal struggle is helpful for the clients.
If we merge results from these researchers and my study, we find that clients in caregiving organizations face a significant risk of being left outside new collaborations.
When new collaboration projects are launched, there is a high probability that cooperation becomes a struggle.

Concluding remarks
In this article, I analysed the "youth care", which is an important part of social work, by describing the significant actors within the empirical field of social work and "youth care" as a salient area within social work. Taking care of presumably troublesome or hopeless youth is indeed a classic task for social work.
"Collaboration" is also an important aspect of social work, and is generally portrayed as being beneficial to authorities and organizations. Previous collaborative research shows that conflicts are common between organizations and authorities that are supposed to cooperate, as well as between collaborating actors. However, what takes place when different actors in the collaboration meet in practice? And how to analyse this?
The analytical discoveries described in my previous work (Basic, 2012) are transferred from an analysed empirical example (interview quotes or field notes) to analytical findings linked to a case (project), and then retransferred to analytic findings in an overall collaboration context, which is represented by earlier research conducted on the social phenomenon of "collaboration". The researcher who carefully analyses single cases must strive to learn something and understand the phenomenon (Ragin and Becker, 1992). It is often problematic in qualitative studies to go from an exhaustive analysis of empirical instances to an overall picture of the context or phenomenon in which all instances, taken together, may be looked upon as a case. Years of close engagement with the data may interfere with the analyst's potential and capacity to contextualize a study more broadly and theoretically, and detailed knowledge about a range of situations in the field may make identifying novel contextualizations more difficult.
This article presents a strategy on how to overcome such difficulties with the help of examples from a study about a "collaboration" project in Swedish youth care. In the cases I have read about in other qualitative studies: a) their findings have been transferred to my case, and b) I hope that my case will be transferable to new empirical findings.
Although the study yielded detailed knowledge about numerous forms and constellations of conflicts within this "collaboration" project, a complete picture of the result did not emerge until all of the instances were reanalysed in terms of previous research on partnerships in additional areas and among other organizations. Parallels and differences between the retold conflicts during interviews and the interactional, in-situation-shaped brawls now revealed that: (1) the "collaboration" project became a fight in Georg Simmel's (1908/70) meaning, and that (2) the client in human service organizations faces an important risk of falling outside such new "collaborations".
This article attempts to specify how this result was realized, and consequently, what it means in practice to examine a set of qualitative data from a wider angle. The analysis in Basic (2012) highlights some overall tendencies that could be taken into consideration when analysing collaboration (and conflicts) between organizations, as well as interpersonal cooperations (and struggle) within and outside the boundaries of juvenile care (Fangen, 2004(Fangen, /2005Becker, 1970: 41-42).